Despite significant advances in our understanding of degenerative diseases of the brainstem and retina, the pathways that underlie most of these disease processes remain undefined. Furthermore, the basis for cell-type specific degeneration and neuron cell death in many important neurological and retinal diseases is unknown. With these issues in mind, we hypothesized that delineation of the molecular basis of the neurological mouse mutant, Purkinje cell degeneration (pcd), would yield insights into how and why certain types of neurons degenerate. pcd is an autosomal recessive mutation that produces a non-developmental degeneration of cerebellar Purkinje cells and retinal photoreceptors. To determine the basis of the degeneration in pcd, we mapped the pcd gene defect to a 1 cM region on chromosome 13 and then identified mutations in the Nna1 gene in two alleles of pcd. Nna1 is a predicted 1218 amino acid protein that has a zinc carboxypeptidase domain with highly conserved orthologues in human (NNA1) and Drosophila (NnaD). Additional homologues of Nna1 exist in mouse, human, moth, and worms, all of which retain the carboxypeptidase domain. Using the results and resources that we have generated, we now wish to advance our understanding of how loss of Nna1 function leads to cell-type specific neurodegeneration in pcd, and thereby determine if Nna1's pathway of action has implications for other diseases. To accomplish these goals, we will begin by confirming that Nna1 is the causal gene by attempting transgenic rescue of the entire pcd phenotype and of one selected neuronal population, namely Purkinje cells. We will seek the mutations responsible for remaining pcd alleles. We will study the expression characteristics of the Nna1 gene product, and will define the nature of neuronal death in pcd mice. We will evaluate Nna1 for carboxypeptidase activity and will determine if retention of carboxy-peptidase activity is required for successful transgenic rescue. To identify Nna1's pathway of action, we will use lines of Drosophila that carry NnaD loss-of-function alleles to perform modifier screens to identify NnaD interacting genes. We will characterize Nna1 interacting proteins detected in a yeast two-hybrid screen, and will compare the results of these two screens to guide future studies.